Caffeine and related methylanthines are behaviorally-active drugs used clinically as respiratory stimulants, but biochemical mechanisms that mediate the behavioral and respiratory effects remain undefined. Studies have implicated both antagonism of adenosine and inhibition of phosphodiesterase as possible mechanisms that mediate their central effects. Because of the widespread use of methylxanthines as drugs and as constituents of good' studies are proposed to (1) investigate in a nonhuman primate model mechanisms that mediate the respiratory-stimulant and behavioral effects of methylxanthines. (2) characterize respiratory changes associated with operant behavior and the modification of those changes by methylxanthines, and (3) determine the respiratory and behavioral effects of caffeine administered chronically. Ventilation (minute volume, tidal volume and respiratory frequency) in unanesthetized rhesus monkeys will be continuously monitored and measured by enclosing the monkey's head in a fitted Plexiglas helmet while a pressure transducer measures differences in pressure produced by inspirations and expirations against a constant air-flow through the helmet. Drug effects on ventilation will be determined during exposure to normal atmospheric conditions and during exposure to elevated concentrations of carbon dioxide in inspired air. A medical gas analyzer will continuously monitor carbon dioxide tension in expired air during all respiration experiments. A wide range of doses of caffeine and selected methylxanthines that differ in their relative affinities for adenosine-receptor subtypes and in their potencies as adenosine antagonists or as phosphodiesterase inhibitors will be administered alone and in combination with adenosine agonists. Operant behavior will be studied while ventilation is being monitored continuously to characterize respiratory changes associated with behavioral activity, and to characterize interacting effects between schedule-controlled behavior and methylxanthines on respiration. Subsequently, caffeine will be administered chronically by scheduling access to water bottles containing drug solution. The behavioral and respiratory effects of chronic administration, including altered sensitivity to other stimulants and to adenosine agonists, will be determined during and after chronic treatment. The technique and method used to measure respiration are proposed as a model for studying basic mechanisms that mediate the central actions of drugs of abuse.